On Wednesday 29 October 2003 07:30 am, robin wrote:
An analogy is website design. I write, or at least try to write, standards-compliant HTML/CSS. If it comes out fine in Mozilla, I can be pretty damn sure it'll look OK in Opera or Konqueror. I also know that there is a risk that Internet Explorer will mangle my code, so from time to time I test my pages in IE. I can't be bothered to spend a lot of time on workarounds to makesure my pages look exactly right on IE, but I at least want them to be legible (same applies to things like Netscape 4.*). I can't afford to take the view that any problem is Microsoft's fault and not my problem, because 90% of people viewing my pages will be using IE.
Of course, in your view, you test on software that is available for free. So, when IE becomes non-standalone with the next version and the only way to obtain the newest version is to purchase the entire Windows OS for $$$, is your view going to remain the same about testing on IE? IOW, are you going to buy the OS and a box to install it on in order to test your web design? And, BTW, you will need a dedicated box because the palladium, er trusted computing initiative stuff will guarantee that you need the entire box dedicated to Windows.
In that case, I'd email a friend and say "Can you check this website?"
The same applies to LG CDROMs. I don't know how common they are where the developers live, but here in Turkey, they are the most common drives, because they're cheap, and actually more reliable than their main rival, Sony. Whateverthe problems, the kernel should have been tested on them.
The reason that standards exist is to eliminate the need for developers to buy one of each different type of drive and test it. That is really the only way to insure compatibility. On the other hand, it would have been trivial for LG to test each of their drives on Linux to insure compatibility and if they did not do so, it is up to the user who chooses to install Linux on his hardware to test and insure that it is okay before they commit to it.
I already mentioned that LG are culpable. As for the responsibility of the user, it is not fair to the user to expect him or her to retest hardware that has worked successfully with previous versions of Linux. We expect that hardware my not work under a new version (for a while I had to keep an old kernel on my system for this very reason) but we do not expect the hardware to be damaged (well, not since the days when X could blow up your monitor!).
At the very least, you should purchase a drive that either others have used before, or that is built to published (and compatible) standards and then hold the manufacturer culpable if those standards are not correct.
Buyer beware is a VERY small price to pay for what we have been given with Linux. This endless bitching about someone else not assuming all responsibility for what each of us do ourselves is really starting to wear thin on me.
The outraged reactions to anyone criticising Linux are starting to wear thin on me.
For anyone that doesn't want to take responsibility for learning, knowing and understanding their own hardware and software, there is an alternative. It is called Windows.
Sure, keep on with this attitude and help ensure the safety of the MS monopoly. In any case, we all know that if you don't want to have to learn about your hardware and software, the alternative is a Macintosh ;-)
[snip]
Be careful what you wish for.. if Mandrake were to adopt a position of responsibility for LG's error, you would unlikly see anything "cutting edge" in mandrake till long after someone else has done it first...
As I said, the fault at the Linux end is really with the kernel developers more than with Mandrake.
Actually, I disagree. The fault is with a manufacturer that deviated from known and published standards, and then failed to notify their customers that they had done so, and in doing so, created the potential for disaster.
Note what I said: "the fault at the Linux end". We are all in agreement about the primary fault lying with LG.
I really don't get the point back to Linux kernel developers. How could they have ever expected any hardware manufacturer to do something as stupid as match up a firmware upgrade command to something as common as a clear buffer command? Given that they are volunteering their time to all of us, why don't some of the current crop of complainers pony up their own resources and start sending hardware to the developers so that testing can be done? Instead of pointing out problems, start providing solutions.
Pointing out problems is a valuable form of support, or would be if people weren't so defensive. When I file a report in bugzilla, I don't expect someone to mail me back and say, "OK, smartypants, where's the patch then?"
To be honest, CDROM's being as cheap as they are now days.. throw the dud one out .(or send it back to LG) and go and get yourself a nice new burner or DVD player (or both) and get on with life....
That's fine if we're talking about a home user with a decent salary. If we're talking about a school in a developing country with a load of CDROMs to replace, I imagine the techie who suggested switching over to Linux isproabbly looking for a job now.
Given the difference in licensing costs on software, I imagine that pointing out that the loss of a bunch of $12 drives not exceeding the cost of paying some $200+ per computer (for windows) would help deflect that particular criticism.
Not if they've already paid for the licences.
Sir Robin
-- "I declare this sentence a performative!"
Robin Turner IDMYO Bilkent Univeritesi Ankara 06533 Turkey
www.bilkent.edu.tr/~robin
Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
